[llvm] [AMDGPU] Switch V_CNDMASK operands to shrink it into VOP2 (PR #135162)

[Matt Arsenault via llvm-commits]

================
@@ -831,6 +836,109 @@ bool SIShrinkInstructions::tryReplaceDeadSDST(MachineInstr &MI) const {
   return true;
 }
 
+unsigned SIShrinkInstructions::getInverseCompareOpcode(MachineInstr &MI) const {
----------------
arsenm wrote:

I thought we already had that? Isn't this equivalent to getting the commuted opcode? 

https://github.com/llvm/llvm-project/pull/135162